Table 4 A comparative analysis of the performance of machine learning models and the human predictive factor (HPF).
Model | Validation accuracy (%) | Test accuracy (%) | Advantage | Disadvantage |
|---|---|---|---|---|
Fine Tree | 54.1 | 83.1 | Easy to interpret decision rules | Over-learning, limited generalisability |
Kernel-based | 55.4 | 81,9 | Handling non-linear relationships | Sensitive to outliers |
kNN | 62 | 100 | Simplicity, efficiency on small data sets | Significant over-learning |
HPF | 62.81 | 72 | Simple, fast to implement | Lower accuracy than machine learning models |
